Quit with an error status on SIGTERM during migration

This prevents the supervisor from thinking that the migration completed
successfully.

In order to do this, I've introduced a new lock around the start (and
finish) of the migration so that we avoid a race between the signal
handler in the server_accept loop and the control thread mirror startup.
Without that, we'd risk successfully starting a migration after the
SIGTERM handler fired, which would be Bad.
This commit is contained in:
Alex Young
2012-10-04 14:41:55 +01:00
parent ddc57e76d1
commit f3e0d61323
10 changed files with 226 additions and 37 deletions

View File

@@ -121,11 +121,12 @@ class Environment
def run_fake( name, addr, port, sock=nil )
fakedir = File.join( File.dirname( __FILE__ ), "fakes" )
fake = Dir[File.join( fakedir, name ) + "*"].sort.find { |fn|
fakeglob = File.join( fakedir, name ) + "*"
fake = Dir[fakeglob].sort.find { |fn|
File.executable?( fn )
}
raise "no fake executable" unless fake
raise "no fake executable at #{fakeglob}" unless fake
raise "no addr" unless addr
raise "no port" unless port

View File

@@ -0,0 +1,19 @@
#!/usr/bin/env ruby
# Wait for a sender connection, send a correct hello, then sigterm the
# sender. We expect the sender to exit with status of 6, which is
# enforced in the test.
require 'flexnbd/fake_dest'
include FlexNBD
addr, port, pid = *ARGV
server = FakeDest.new( addr, port )
client = server.accept( "Timed out waiting for a connection" )
client.write_hello
Process.kill(15, pid.to_i)
client.close
server.close
exit 0

View File

@@ -0,0 +1,20 @@
#!/usr/bin/env ruby
# Connect to the listener, wait for the hello, then sigterm the
# listener. We expect the listener to exit with a status of 6, which
# is enforced in the test.
require 'flexnbd/fake_source'
include FlexNBD
addr, port, pid = *ARGV
client = FakeSource.new( addr, port, "Timed out connecting." )
client.read_hello
Process.kill( "TERM", pid.to_i )
sleep(0.2)
client.close
exit(0)

View File

@@ -136,11 +136,11 @@ class ValgrindKillingExecutor
def call( err )
Process.kill( "KILL", @pid )
$stderr.puts "*"*72
$stderr.puts "* Valgrind error spotted:"
$stderr.puts err.to_s.split("\n").map{|s| " #{s}"}
$stderr.puts "*"*72
Process.kill( "KILL", @pid )
exit(1)
end
@@ -323,7 +323,8 @@ module FlexNBD
def serve( file, *acl)
run_serve_cmd( serve_cmd( file, acl ) )
cmd = serve_cmd( file, acl )
run_serve_cmd( cmd )
end
def listen(file, *acl)

View File

@@ -28,6 +28,11 @@ class TestDestErrorHandling < Test::Unit::TestCase
end
def test_sigterm_has_bad_exit_status
@env.nbd1.can_die(1)
run_fake( "source/sigterm_after_hello" )
end
def test_disconnect_after_hello_causes_error_not_fatal
run_fake( "source/close_after_hello" )
assert_no_control

View File

@@ -19,12 +19,24 @@ class TestSourceErrorHandling < Test::Unit::TestCase
end
def expect_term_during_migration
@env.nbd1.can_die(6,9)
end
def test_failure_to_connect_reported_in_mirror_cmd_response
stdout, stderr = @env.mirror12_unchecked
expect_term_during_migration
assert_match( /failed to connect/, stderr )
end
def test_sigterm_after_hello_quits_with_status_of_1
expect_term_during_migration
run_fake( "dest/sigterm_after_hello" )
end
def test_destination_hangs_after_connect_reports_error_at_source
run_fake( "dest/hang_after_connect",
:err => /Remote server failed to respond/ )
@@ -36,6 +48,7 @@ class TestSourceErrorHandling < Test::Unit::TestCase
:err => /Mirror was rejected/ )
end
def test_wrong_size_causes_disconnect
run_fake( "dest/hello_wrong_size",
:err => /Remote size does not match local size/ )
@@ -43,38 +56,45 @@ class TestSourceErrorHandling < Test::Unit::TestCase
def test_wrong_magic_causes_disconnect
expect_term_during_migration
run_fake( "dest/hello_wrong_magic",
:err => /Mirror was rejected/ )
end
def test_disconnect_after_hello_causes_retry
expect_term_during_migration
run_fake( "dest/close_after_hello",
:out => /Mirror started/ )
end
def test_write_times_out_causes_retry
expect_term_during_migration
run_fake( "dest/hang_after_write" )
end
def test_rejected_write_causes_retry
expect_term_during_migration
run_fake( "dest/error_on_write" )
end
def test_disconnect_before_write_reply_causes_retry
expect_term_during_migration
run_fake( "dest/close_after_write" )
end
def test_bad_write_reply_causes_retry
expect_term_during_migration
run_fake( "dest/write_wrong_magic" )
end
def test_pre_entrust_disconnect_causes_retry
expect_term_during_migration
run_fake( "dest/close_after_writes" )
end